Answer Generation through Unified Memory over Multiple Passages

• Published in: Transactions of the Japanese Society for Artificial Intelligence Vol. 37; no. 4; pp. C-LC4_1 - 9
• Main Author: Nakatsuji, Makoto
• Format: Journal Article
• Language: Japanese; English
• Published: Tokyo The Japanese Society for Artificial Intelligence 01.07.2022; Japan Science and Technology Agency
• Subjects: answer generation; Coders; Encoders-Decoders; machine reading comprehension; Questions
• ISSN: 1346-0714; 1346-8030
• DOI: 10.1527/tjsai.37-4_C-LC4

Machine reading comprehension methods that generate answers by referring to multiple passages for a question have gained much attention in AI and NLP communities. The current methods, however, do not investigate the relationships among multiple passages in the answer generation process, even though topics correlated among the passages may be answer candidates. Our method, called neural answer Generation through Unified Memories over Multiple Passages (GUM-MP), solves this problem as follows. First, it determines which tokens in the passages are matched to the question. In particular, it investigates matches between tokens in positive passages, which are assigned to the question, and those in negative passages, which are not related to the question. Next, it determines which tokens in the passage are matched to other passages assigned to the same question and at the same time it investigates the topics in which they are matched. Finally, it encodes the token sequences with the above two matching results into unified memories in the passage encoders and learns the answer sequence by using an encoder-decoder with a multiple-pointer-generator mechanism. As a result, GUM-MP can generate answers by pointing to important tokens present across passages. Evaluations indicate that GUM-MP generates much more accurate results than the current models do.